deps: update mlx-lm

Summary of changes:

DSV3 MLA
Fix sliding window mask during generation
Fix batch mamba
Fix Step 3.5 Flash model conversion
Deepseek V3.2 implementation fixes
fix: handle GLM 4.7 tool call fallbacks
server: support chat_template_kwargs and top_logprobs
Add Step 3.5 Flash
allow creation of BatchRotatingKVCache instead of BatchKVCache when empty cache(s) are passed to BatchGenerator
enable loading custom models
fix cli
Support distributed inference in the server
fix mixed quant
Add LongCat Flash Lite
actually add cli
Fix Kimi K2.5 tool call handling
Fix for Exception - MultiLinear.to_quantized() missing 'mode'
Fix NemotronH config compatibility with HuggingFace format
Bump mlx version and version

Full changelog: 96699e6d...f18526f8

pyproject.toml

@@ -19,7 +19,7 @@ dependencies = [
     "anyio==4.11.0",
     "mlx==0.30.4; sys_platform == 'darwin'",
     "mlx[cpu]==0.30.4; sys_platform == 'linux'",
-    "mlx-lm",
+    "mlx-lm==0.30.6",
     "tiktoken>=0.12.0", # required for kimi k2 tokenizer
     "hypercorn>=0.18.0",
     "openai-harmony>=0.0.8",
@@ -63,7 +63,6 @@ members = [
 
 [tool.uv.sources]
 exo_pyo3_bindings = { workspace = true }
-mlx-lm = { git = "https://github.com/ml-explore/mlx-lm", branch = "main" }
 # Uncomment to use local mlx/mlx-lm development versions:
 # mlx = { path = "/Users/Shared/mlx", editable=true }
 # mlx-lm = { path = "/Users/Shared/mlx-lm", editable=true }

uv.lock

@@ -415,7 +415,7 @@ requires-dist = [
     { name = "mflux", specifier = "==0.15.4" },
     { name = "mlx", marker = "sys_platform == 'darwin'", specifier = "==0.30.4" },
     { name = "mlx", extras = ["cpu"], marker = "sys_platform == 'linux'", specifier = "==0.30.4" },
-    { name = "mlx-lm", git = "https://github.com/ml-explore/mlx-lm?branch=main" },
+    { name = "mlx-lm", specifier = "==0.30.6" },
     { name = "openai-harmony", specifier = ">=0.0.8" },
     { name = "pillow", specifier = ">=11.0,<12.0" },
     { name = "psutil", specifier = ">=7.0.0" },
@@ -1072,8 +1072,8 @@ wheels = [
 
 [[package]]
 name = "mlx-lm"
-version = "0.30.5"
-source = { git = "https://github.com/ml-explore/mlx-lm?branch=main#96699e6dadb13b82b28285bb131a0741997d19ae" }
+version = "0.30.6"
+source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "mlx", marker = "sys_platform == 'darwin'" },
@@ -1083,6 +1083,10 @@ dependencies = [
     { name = "sentencepiece", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "transformers", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/76/cb/815deddc8699b1f694d7e1f9cbed52934c03a8b49432c8add72932bb2f0b/mlx_lm-0.30.6.tar.gz", hash = "sha256:807e042d7040268f1b19190b7eaefd8b2efbff5590a65460974ad4225b91dda1", size = 271733 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/20/5f/01d281f1fa8a1521d5936659beb4f5ab1f32b463d059263cf9d4cef969d9/mlx_lm-0.30.6-py3-none-any.whl", hash = "sha256:a7405bd581eacc4bf8209d7a6b7f23629585a0d7c6740c2a97e51fee35b3b0e1", size = 379451 },
+]
 
 [[package]]
 name = "mlx-metal"